write-documentation
Skill2 starsChanged 15 days agoMITView on GitHub
Use when creating, updating, or reviewing human-facing documentation, including guides, READMEs, API references, runbooks, and decision records.
We split this skill into the rules it gives an agent
11 rules. 0 are common, 10 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
10
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Create documentation that lets its intended reader understand or complete a task from accurate, findable information.
- Distinctive: line 21
Read the complete target, relevant neighboring pages, navigation, and writing conventions.
- Distinctive: line 26
Check behavior, defaults, commands, limits, and terminology against the relevant code, configuration, tests, approved decisions, or current …
- Distinctive: line 28
Keep current behavior distinct from proposals, planned releases, historical decisions, and unknowns.
- Distinctive: line 35
Use reader needs to choose the structure: guided learning, completing a known task, looking up a fact, or understanding an explanation.
- Distinctive: line 40
Read references/document-types.md for the relevant document's checks.
- Distinctive: line 47
Follow the project's language and framework.
- Distinctive: line 52
Use screenshots, diagrams, or tables when they clarify a state, relationship, or comparison.
- Distinctive: line 59
Follow the documented path your change affects, in order, using safe local fixtures or an appropriate test environment within the active …
- Distinctive: line 72
Read the result without relying on private conversation context.
- Distinctive: line 81
Update affected navigation, local links, and examples within scope.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Write documentation
Create documentation that lets its intended reader understand or complete a task
from accurate, findable information. communicate-clearly owns prose and voice; load
it when available and not already active. This skill owns documentation
structure, source accuracy, examples, and validation. agent-instructions owns
instructions for agents.
Find the reader, source, and destination
Establish the reader, their task and prior knowledge, the output language, and the product version or state being described. Infer these from the request and existing documentation; ask only about a material unresolved choice. A small correction does not need a planning interview.
Read the complete target, relevant neighboring pages, navigation, and writing conventions. Locate the canonical source before creating another page. Preserve useful existing structure, generated-file ownership, and user edits. Make review findings or a draft when requested; edit files only within the authorized scope.
Check behavior, defaults, commands, limits, and terminology against the relevant code, configuration, tests, approved decisions, or current primary documentation. Keep current behavior distinct from proposals, planned releases, historical decisions, and unknowns. Code can establish behavior without proving why it was chosen or whether it is approved policy. Surface consequential contradictions instead of silently resolving them in favor of whichever source is newest.
Organize for the task
Use reader needs to choose the structure: guided learning, completing a known task, looking up a fact, or understanding an explanation. These Diataxis distinctions can apply to sections. Keep a useful mixed README or existing page intact rather than forcing a new file for every type.
Read references/document-types.md for the relevant document's checks. Lead with what the page lets this reader do or understand. Place prerequisites and consequential conditions before the actions they govern. Use descriptive headings, ordered steps where order matters, and links to deeper explanation when it would interrupt the task. Include enough context to understand the step; a how-to guide need not be unexplained commands.
Follow the project's language and framework. Examples should demonstrate the documented contract with the smallest sufficient setup. Distinguish runnable code from illustrative fragments and name any omitted context. Preserve identifiers, placeholders, required markup, units, and version-specific behavior.
Use screenshots, diagrams, or tables when they clarify a state, relationship, or comparison. Verify that their labels and depicted behavior match the relevant version; provide meaningful text alternatives. Assets are supporting evidence, not proof that the instructions work.
Verify what the reader will use
Follow the documented path your change affects, in order, using safe local fixtures or an appropriate test environment within the active authorization. Check prerequisites, commands, inputs, expected output, and how the reader recognizes success. Run applicable documentation build, link, schema, and example checks from the project. Use its established generator for generated documentation instead of editing derived output.
A documentation request does not authorize a production migration, deployment, message, or destructive operation just because a guide contains its command. Validate such steps in an authorized disposable environment or report what remains unverified. Distinguish static inspection, compilation, execution, and observed application behavior; do not infer one from another or manufacture passing evidence.
Read the result without relying on private conversation context. Can a reader with the stated prerequisites find the entry point, supply the required inputs, follow the path, and recognize the outcome? For complex or high-consequence guides, an independent reader check can expose hidden assumptions when available and authorized. Treat model feedback as a check of this sample, not proof of universal comprehension. Repair gaps and rerun the affected checks.
Integrate and finish
Update affected navigation, local links, and examples within scope. When moving a published page, account for existing links and the site's redirect mechanism; do not move it merely to impose a preferred taxonomy. Keep requested language versions aligned on meaning and behavior, adapting natural wording rather than translating the source language's syntax mechanically. Report any unupdated versions and why.
Documentation maintenance does not by itself create a new docs platform, glossary, CI workflow, policy file, or publication step. Use established owners and conventions and propose a separate material change when one is actually needed.
Deliver the requested artifact or review. For file changes, report the paths, material content changes, exact checks and outcomes, and unresolved gaps. Finish when the content serves the reader's task, source claims are supported or explicitly qualified, applicable checks pass, and any unavailable validation is stated.
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- r3bl-org/r3bl-open-core★ 486same name
- Indium-AI-Labs/indium-agentkit★ 18same name
- Ionate-Technologies/indium-agentkit★ 17same name
- Yassimba/loom★ 9same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository48 other files it gives its agents.
- accessibilitySkill
- agent-instructionsSkill
- analyze-change-effectsSkill
- article-processingSkill
- ci-cd-automationSkill
- comment-styleSkill
- commitSkill
- communicate-clearlySkill
- compare-solutionsSkill
- create-project-instructionsSkill
- daily-meeting-updateSkill
- debugSkill
- deprecate-and-removeSkill
- design-code-structureSkill
- domain-searchSkill
- error-handlingSkill
- explain-codeSkill
- explain-decisionsSkill
- github-actionsSkill
- handoffSkill
- humanizeSkill
- issue-workflowSkill
- observabilitySkill
- people-memorySkill
- performance-optimizationSkill
- planningSkill
- pr-followupSkill
- prototypeSkill
- pr-review-followupSkill
- pr-reviewSkill
- prSkill
- report-work-activitySkill
- retroSkill
- review-code-changesSkill
- simplify-codeSkill
- skill-maintenanceSkill
- stacked-prSkill
- task-breakdownSkill
- tddSkill
- teachSkill
- typescript-best-practicesSkill
- verification-authoringSkill
- verifySkill
- wayfinderSkill
- workflow-to-skillSkill
- work-modeSkill
- worktrunkSkill
- youtube-processingSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

